This is my first ML project. Wish me luck.
My aim is to create a simple python script that open live camera then giving a sign wether the mask is being used or not. My plan is to use tensorflow model and opencv to predict and detect faces. Perhaps I could put my work into a website so that people can try what I have done.
So far I have created
- Machine Learning model that can detect people wearing their mask or not. I am using EfficientNetB7 mainly for the convolutional part of my model. And I am getting a pretty good accuracy with the model. But the size is too big and the parameters are to many. I also opted to use MobileNetv2 if I needed smaller model. I create the model using google colab.
- OpenCV script that can detect faces. I am using Haar Cascade method to detect faces.
- Script that can open the computer's webcam then detect faces + mask on or off.
To Do List:
- Put the whole thing into a website. But I still don't know how to do this.